Econbrowser

Last week I reported that the new GDP figures released by the Bureau of Economic Analysis led to a slight increase in my GDP-based recession probability index from 3.4% to 4.7%. Pro-Growth Liberal of Angry Bear, in comments placed at the blogs of Economist's View and William Polley, expressed amusement that the same GDP figures led him, a pessimist, to become more optimistic, and led me, an optimist, to become more pessimistic.

Prompted by the reaction of many others to the new GDP figures, I was curious to take a further look at why the recession probability index was reacting the way it did. Argmax.com in particular emphasized that part of last week's news release included downward revisions to the estimates of GDP for a number of earlier quarters, and my algorithm uses the complete new history as released last week to form an assessment about where things stood in the first quarter. Out of curiosity, I tried the experiment of just adding a 3.4% growth rate for 2005:II without changing any of the other numbers as they had been reported last May, and found that, if there had been no revisions to earlier data, the index would only have risen to 4.3 rather than 4.7. So BEA's downward historical data revisions explain part of the reason I was coming up with a slightly pessimistic reading of the new GDP data. But the biggest factor remains the one I mentioned last week, that 3.4% is actually a bit slower annual growth rate than we typically observe during an economic expansion. Nevertheless, I was perhaps negligent in not emphasizing last week that a value of 4.7 for the index still represents a very, very favorable economic outlook.

There was quite a range of other interesting responses to the new GDP figures around the blogosphere, including Angry Bear, Tim Duy, William Polley, General Glut, the Big Picture, Lakeshore Laments, and Capital Spectator. I found all of these very informed and informative, but the colorful variety of responses persuaded me all the more that there was also some value added in what I attempted, which was to provide a purely objective summary of the numbers. And the nature of that summary is, if you look just at the GDP figures themselves, ignoring the individual components, the economy appears to be chugging along nicely, though the overall effect of the new figures, including the historical revisions, would be to make an objective observer ever so slightly more bearish than he or she might have been before seeing the numbers.

But while I was playing statistician, others were busy being real economists, finding some very important information in the individual components that make up the GDP totals. Tim Duy in particular, as well as William Polley and Angry Bear, noted that inventory reductions were contributing a net drag on GDP that amounted to -2.3% growth at an annual rate, meaning that, if inventory additions had continued at their previous pace rather than fallen, real GDP would have grown at a 5.7% annual rate in the second quarter.

Changes in inventories of course should not just be ignored-- any sales that come out of inventory mean no new income generated for those who produce the goods, and in many cases the economic downturn associated with a recession can be entirely attributed to precisely this reality. However, inventories are often very important to examine because they can help predict where GDP is likely to head next. In the present case, if inventory additions in 2005:III just return to the average value we've seen over the last year, that alone would kick in an extra 2% to the GDP annual growth rate calculated for next quarter. So, even if final sales are fairly lackluster from here, we should still get a good value for GDP growth for the coming quarter.

This cheery outlook is also not without some concerns of its own, however. The Big Picture notes that a huge chunk of the inventory drop was due to auto sales incentives on which GM managed to lose $1.2 billion in the second quarter. Tim Duy helped me track down these inventory figures (from BEA Table 5.6.6B), which show that the change in automobile inventory investment accounted for 38% of the total national change in inventory investment in 2005:II, contributing -0.9% to GDP growth just from the auto sector. GM's losses led me to worry earlier that big production cutbacks from the U.S. automakers could be coming soon

Tim nevertheless argues that GM and Ford are less important for the U.S. economy than they used to be. Although I agree that the day is coming when they will not be that critical (and GM is perhaps doing its part to precipitate the arrival of that date), I'm not convinced that it's here yet.

Angry Bear noted that the GDP figures also show a decrease in the contribution of consumption and increased contribution of net exports. With exports increasing and imports decreasing in 2005:II compared to 2005:I, net exports contributed +1.6% to the second quarter annual GDP growth rate. It would surely be far healthier for U.S. economic growth to be led by exports rather than by consumption spending. I've been watching to see if any of this might bring some more cheer to Brad Setser's gloom, but so far, it seems not.

In addition to all this, last week we received news that housing construction remained very strong in June. Taken together, these details lead me to lighten some of the pessimism about housing that I'd expressed before these data came out. My reasoning had been that since lower interest rates produced the house price increases, rising interest rates could easily set that same process into reverse. The new data lead me to conclude that the economy should be able to withstand a few more rate hikes before that happens, which is just as well, since that's doubtless what the Fed is going to do anyway. I nevertheless continue to urge the Fed to keep watching autos, housing, and the yield curve with great caution.

But I can't allow Angry Bear to be more of a Cheerful Optimist than I am, can I?